| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Division of Maternal-Fetal Medicine,
b Reproductive Biology Unit,
c Division of Reproductive Medicine,
d Department of Obstetrics and Gynecology, University of Ottawa/The Ottawa Hospital, Ottawa Health Research Institute, The Ottawa Hospital, Ottawa, Ontario, Canada K1H 8L6
ABSTRACT
Maternal smoking is associated with severe perinatal complications and significant placental pathologies with underlying ultrastructural changes. In this study, we examined the influence of maternal smoking on trophoblast apoptosis throughout development and correlated those findings with changes in expression of X-linked inhibitor of apoptosis protein (Xiap) as well as Fas and Fas ligand (FasL). Trophoblast apoptosis was determined by DNA fragmentation and TUNEL. Protein expression was assessed by Western blotting and immunohistochemistry. Maternal smoking was associated with increased trophoblast apoptosis in the first trimester but decreased trophoblast apoptosis near term. Placental Xiap levels decreased significantly throughout development in nonsmokers (P < 0.05) but remained elevated in smokers. Fas and FasL levels did not vary significantly throughout development nor between groups. However, procaspase-3 levels were significantly increased in smokers at term. Our results suggest that maternal smoking has different effects at different stages of trophoblast differentiation and that this is regulated in part through modulations in placental Xiap expression.
apoptosis, environment, placenta, pregnancy, trophoblast
INTRODUCTION
Maternal smoking is associated with several perinatal complications, including spontaneous abortion [1], reduced fetal growth [2, 3], prematurity [3], and increased perinatal mortality and morbidity [4]. Investigations on the influence of smoking on placental morphology and function have begun to define the pathophysiological mechanisms involved. Ultrastructural changes including cytotrophoblast hyperplasia, thickening of the trophoblast basement membrane, and decreased capillary density at the terminal villi are believed to contribute to reduced placental nutrient and oxygen transfer [5–7]. More recent data have also implicated a direct effect of smoking on cytotrophoblast proliferation and differentiation in the first trimester of pregnancy [8, 9]. In addition, disturbances in placental function as evidenced by decreased amino acid uptake and transfer have also been implicated [10]. Although available data suggest that maternal smoking influences placental growth, development, and function, all of which are important determinants of fetal growth and well-being, most studies to date have focused on only a specific developmental phase, and systematic information throughout gestation remain lacking.
Apoptosis is a physiologic form of cell death and is important in the control of cell population. It plays key roles in the regulation of various physiological and pathological conditions, including vertebrate development, ovarian follicular atresia, immune disorders, and cancers [11–15]. Apoptosis is present in the placenta throughout gestation, increases near the end of gestation, and is believed to be physiologically important for normal placental growth and development [16, 17]. Increased trophoblast apoptosis has been documented in placentae of growth-restricted fetuses, and maternal smoking is associated with decreased placental apoptosis at term [18, 19]. These findings are consistent with in vitro findings that have shown an inhibitory effect of nicotine on apoptosis [20]. However, placentae from earlier developmental stages have not been examined, nor have the underlying mechanisms involved in the regulation of apoptosis. Although the Fas/Fas ligand (FasL) system is expressed in the human placenta throughout gestation, its role in the regulation of placental apoptosis remains unclear [21, 22]. Our prior studies have shown that X-linked inhibitor of apoptosis protein (Xiap) contents decreased significantly near term, coincidental to increased trophoblast apoptosis, suggesting that this protein may play an important role the regulation of apoptosis in the placenta [16]. Indeed, Xiap is an intracellular survival protein and is known to modulate the activation and activities of caspases, a family of cell death proteases that are important in apoptotic signaling [23]. However, no information has been available to date on the influence of maternal smoking on placental expression of this protein.
In the present study, we have compared the extent of placental apoptosis between nonsmokers and smokers during the first, second, and third trimesters of gestation. We have also examined the changes in Xiap and Fas/FasL expression to determine whether these cell survival and death proteins play a role in the dysregulation of placental apoptosis in mothers who smoke during pregnancy.
MATERIALS AND METHODS
Sample Collection
Placental tissues were obtained from 53 first-trimester and 20 second-trimester therapeutic terminations, and 56 third-trimester pregnancies immediately at delivery. Among them, the number of mothers who smoked and did not smoke were 28 and 25, 12 and 8, and 30 and 26 in the first, second, and third trimesters, respectively. All pregnancies were dated according to the last menstrual period or by early ultrasound. Ultrasound assessment during the first and second trimesters confirmed fetal viability and gestational ages.
The exclusion criteria for the study were fetal demise, fetal anomaly, fetal distress, chromosomal abnormality, chorioamnionitis, premature rupture of membrane, and maternal medical complications, including preeclampsia and maternal substance use. Smoking status was established through patient interviews and by systematic verification of the patient's chart. For third-trimester deliveries, maternal age, mode of delivery, gestational age, birthweight, and cord gases were prospectively noted. Informed consent was obtained from all participants and the study was approved by The Ottawa Hospital Research Ethics Board.
Samples were obtained randomly over three to five areas of the placenta and were divided in two portions for subsequent analysis; one was frozen immediately at -70°C and another was fixed in 4% paraformaldehyde.
Assessment of Apoptotic DNA Fragmentation
Apoptotic cell death was assessed on the basis of DNA fragmentation and confirmed through visualization of discrete DNA fragments of 185 base pair multiples on agarose gel electrophoresis. DNA was extracted from each specimen using the Qiagen Tissue Amp Kit (Qiagen Inc., Chatsworth, CA) and end-labeled by incubating with terminal end transferase (Roche Molecular Biochemicals, Laval, PQ, Canada) and [
-32P]ddATP. Briefly, 500 ng DNA sample was added to a mixture (5 µl of 5x TdT buffer, 2.5 µl of 10x CoCl2, 0.5 µl TdT enzyme, and 0.5 µl of 10 mCi/ml [32P]ddATP and AE buffer) to a total volume of 25 µl, and then incubated at 37°C for 60 min. Unincorporated nucleotides were removed with the Qiagen nucleotide removal kit (Qiagen) and the samples were subsequently resolved by 1.8% agarose. The dried gel was then exposed to a BioRad PhosphorImager, and low molecular weight DNA (<6 kilobase pairs) and genomic DNA were densitometrically quantified. The gel was then exposed to x-ray film at -80°C. To correct for possible uneven gel loading, the ratio of low molecular weight DNA (representing apoptosis) to genomic DNA was calculated for each sample and means of ratios in each trimester were compared. Examination of multiple samples for each placenta (e.g., third-trimester specimens) indicated that the variability in apoptosis in different areas within the same placenta was less than 20%. The intraobserver variability, determined by performing two separate DNA ladder analyses on a same sample, was approximately 5%. The intraplacental variability justified the need for examining several areas of a large number of placentae, especially in the third trimester, and the use of the means of these results for comparisons.
Terminal Deoxynucleotidyl Transferase-Mediated dUTP-Biotin End-Labeling
TUNEL was performed to identify the placental cell type or types undergoing apoptosis, using the In Situ Cell Death Detection Kit, POD (Roche Molecular Biochemicals, Laval, PQ, Canada). Mounted sections were stained with hematoxylin and eosin (H&E) for routine histology and examined under light microscopy. Subsequently, adjacent sections were treated (1 min) with microwave in 0.01 M citrate buffer. They were then washed with PBS and incubated with terminal transferase in the presence of fluorescein isothiocyanate (FITC)-conjugated dUTP. Signal was converted with antifluorescence in antibody conjugated with horseradish peroxidase (HRP) and 3,3'-diaminobenzidine (DAB) substrate. Positive controls were generated by incubating with DNase I (1 µg/ml before labeling). Slides were then photographed under light microscopy (400x magnification) and positive cells were identified and counted separately by two blinded investigators. The apoptotic index was calculated for each sample based on the number of TUNEL-labeled trophoblasts per 1000 total trophoblasts. This number was then expressed as a percentage, and the means of each group were compared.
Western Blot
Placental tissues (50–100 mg) were homogenized in RIPA (1x PBS, 1% Nonidet P-40 [Sigma, Oakville, ON, Canada], 0.5% sodium deoxycholate, 0.1% SDS containing protease inhibitor [phenylmethylsulfonyl fluoride 1 mM, aprotinin 10 µg/ml, and leupeptin 10 µg/ml]). The homogenates were sonicated (10 sec) and centrifuged (14 000 x g, 4°C, 20 min). The protein content of the supernatant was determined with the Bio-Rad DC protein assay kit. Aliquots of the supernatant (50 µg) were resolved by SDS-PAGE and electrotransferred to nitrocellulose membranes, including one pooled sample as an internal control in each gel. The membranes were blocked (1 h at room temperature) in blotto (Tris-buffered saline pH 7.6 with 0.05% Tween 20 [TBS-T], 5% dehydrated nonfat milk). After washing in TBS-T (four times for 5 min), the membranes were incubated (1 h at room temperature) with each antibody (FasL [C-178], proliferating cell nuclear antigen [PCNA; Santa Cruz Biotechnology, Santa Cruz, CA], procaspase-3 [Pharmingen, Mississauga, ON, Canada], and Xiap [ApoptoGen Inc., Ottawa, ON, Canada]) diluted in blotto. After washing, the membranes were incubated with human immunoglobulin G (IgG) preabsorbed HRP-conjugated second antibody (1:2000) in blotto for another 1 h, and washed again. Peroxidase activity was visualized with the enhanced chemiluminescence kit according to the manufacturer's instructions. The protein content was determined desitometrically and normalized with an internal control on the same membrane.
Immunohistochemistry
Cubes of placenta fixed in 10% neutral buffered formalin were dehydrated through a graded series of ethanol and embedded in paraffin. Serial sections (4–5 µm) were deparaffinized using xylene and a graded series of ethanol. For FasL and Xiap immunostaining, sections were heated by microwave for 15 min in 0.01 M citrate buffer pH 6.0, and immunostained with the DAKO LSAB Kit (DAKO Corporation, Mississauga, ON, Canada). Briefly, the sections were incubated for 10 min in PBS with 3% H2O2 and after blocking, with Monoclonal Fas antibody (Santa Cruz Biotechnology, clone 13; 1:25), FasL (Santa Cruz Biotechnology, C-178; 1:50), Xiap (Apoptogen 1:50), and PCNA (1:25) for 30 min at room temperature. The samples were incubated with biotinylated second antibodies (30 min), then with streptavidin HRP conjugated and visualized with 3-amino-9 ethylcarbazole (AEC). The specimens were counterstained with hematoxylin. A stain-destain-stain technique was used to assess the TUNEL signal and Xiap expression in the same cells. Briefly, TUNEL was performed first as described above, except that AEC (instead of DAB) was used as a substrate. After TUNEL images were captured, TUNEL-label sections were destained with 1% HCl in 70% ethanol (1 min) to remove residual signals and Xiap immunohistochemistry was then performed in the same section as described above.
Statistical Analysis
Data were logarithmically transformed to remove heterogeneity of variance prior to analysis by two-way ANOVA. Differences between experimental groups were assessed by the Student-Newman-Keuls test. Statistical significance was inferred at P < 0.05.
RESULTS
There were no differences in maternal demographics and in gestational age at termination in the first-trimester group. For the third-trimester group, no differences existed between smokers and nonsmokers in maternal age, gravidity, parity, mode of delivery, gestational age, Apgar scores, and cord pH (data not shown). As expected, mean birthweight was lower (P < 0.05) in fetuses of smokers (2710 ± 158 g) than those of nonsmokers (3302 ± 179 g). Mean daily cigarette consumption in the first and third trimesters were 18 ± 3 and 12 ± 3, respectively.
Apoptosis was evidenced in all samples examined. Quantification of low molecular weight DNA fragmentation in first-trimester placental samples revealed a more significant degree of apoptosis in smokers compared with nonsmokers (P < 0.05). Third-trimester placental sample analysis did not reveal a significant difference in apoptotic DNA fragmentation in smokers when compared with nonsmokers (Fig. 1). However, although no difference in apoptotic DNA fragmentation was present in nonsmokers throughout development, placentae of smokers displayed a statistically significant decrease in apoptosis from first to second and third trimesters (Fig. 1; P < 0.05).
|
TUNEL labeling also demonstrated the presence of apoptotic cells on all placental samples. This was confirmed by cell morphology on H&E stain. Whereas no statistically significant change in the number of TUNEL-positive trophoblasts was seen in normal placentae from first to third trimester, the opposite was true in those exposed to maternal smoking. Indeed, placentae of smokers displayed less apoptosis in the third trimester when compared with the first (mean ± SEM: 4.97% ± 1.19% vs. 11.82% ± 1.09%, respectively). Consequently, in the third trimester, apoptosis was decreased in placentae of smokers compared with nonsmokers (mean ± SEM: 4.97% ± 1.19% vs. 8.45% ± 0.94%, respectively).
Xiap was localized in cytotrophoblasts but more intensely in syncytiotrophoblasts in the first trimester. In the third trimester, Xiap immunosignals were detected primarily in syncytiotrophoblasts. Placentae of smokers in the third trimester displayed more intense and frequent Xiap immunoreactive trophoblasts and less TUNEL-labeled cells when compared with those of nonsmokers (Fig. 2). Xiap immunosignals were very rarely observed in TUNEL-positive cells. PCNA staining, indicative of proliferating activity, was most evident in cytotrophoblasts. Western blot analysis indicated that placental Xiap protein content in nonsmokers decreased markedly as pregnancy progressed to term (P < 0.05). In contrast, Xiap levels in smokers remained high and not significantly different from first trimester samples, thus resulting in a significantly higher placental Xiap level in smokers compared with nonsmokers at term (Fig. 3). The changes in placental Xiap content were associated with reciprocal alterations in trophoblast apoptosis (Fig. 2). Whereas PCNA protein content decreased during normal development and reached its lowest level in the third trimester, it remained unaltered in smokers (Fig. 3).
|
|
Whereas immunosignals for Fas were present mostly in cytotrophoblasts in the first trimester, they were most intense in syncytiotrophoblasts at term. As observed with Xiap, FasL signals were localized primarily in trophoblasts in all trimesters, particularly in syncytiotrophoblasts. The intensities of immunosignals for Fas and FasL did not differ significantly between smokers and nonsmokers during the first and third trimesters (data not shown). Western analyses of the protein extracts also revealed no significant differences in Fas and FasL between first and third trimester and between smokers and nonsmokers (Fig. 3). There were no significant differences in procaspase-3 contents between smokers and nonsmokers in the first two trimesters, although levels were higher in smokers near term (Fig. 3).
DISCUSSION
Maternal smoking has been associated with marked ultrastructural and functional placental changes. First-trimester trophoblasts are known to be actively differentiating cells and more susceptible to noxious stimuli. Cytotrophoblasts from smokers have reduced invasive potential, poor differentiation, less outgrowth and column formation in the first trimester, characteristics that are directly related to the concentration of nicotine [8]. Absence of cytotrophoblast stem cells and abnormal thinning of the syncitium have also been observed [9]. Whereas epidermal growth factor (EGF) is known to protect trophoblasts against TGFß-induced apoptosis, [24] its receptors are down-regulated by benzopyrene, a byproduct in cigarette smoke [25]. However, the physiological basis for the decreased invasion and outgrowth in smokers is unknown. Our present results suggest that apoptosis, a phenomenon that is important for tissue remodeling and removal of damaged or poorly differentiated cells, may account for the observed changes.
Previous studies from our laboratory have demonstrated that placental apoptosis increases with normal development [16]. In the present investigations, we have extended these studies and have shown that this form of cell death decreases in placentae of smokers from first to third trimester, when compared with normal gestational age matched controls. Our results are consistent with current knowledge on placental development and trophoblast differentiation. Defects in trophoblast development and formation as a consequence of smoking are, however, not limited to cytotrophoblasts. These alterations frequently include calcification, thickening of the trophoblast layer, and decreased uptake and transport of amino acids, suggesting that smoking may also have adverse effects on the syncytium [5–7, 10]. Our present results clearly demonstrate that smoking induces apoptosis of first trimester syncytiotrophoblasts and are in accord with a previous report on abnormal thinning of the syncytium in placentae of smokers [9], which may lead to suboptimal placental uptake and transport of amino acids [10]. This is consistent with the association between maternal smoking and fetal growth restriction [2], as a functional placental mass is essential to normal fetal growth.
Interestingly, placentae from smokers in the third trimester displayed a lower degree of trophoblast apoptosis when compared with controls. These results are in accord with a prior study by Marana et al. [19] on third trimester placentae, and support speculations that maternal smoking may be less damaging in the last half of pregnancy [26], probably as a consequence of the highly differentiated state of the cells near term, which confers more resistance to insults. Indeed, it is well accepted that more-mature placentae have significant functional reserve capacity. It is possible that as cellular damage occurs in the first trimester, apoptosis constitutes an important compensatory mechanism through which the removal of these damaged cells allows for the formation of new villi. In addition, nicotine-induced apoptotic trophoblast cell death and benzopyrenene-mediated susceptibility to this process (through down-regulation of EGF receptors) could further contribute to the increase in apoptosis observed. Subsequently, as a result of the significant placental damage incurred early in development and an ongoing exposure to maternal smoking, hypoxic conditions likely prevail in the placenta [5–7], thereby resulting in increased cell proliferation and decreased apoptosis in an attempt to restore placental function.
Precisely how nicotine may influence this process is not known. Nicotine has been shown to inhibit apoptosis in a variety of cells exposed to hypoxia [20, 27, 28]. It has also been shown to increase DNA synthesis and proliferation in arterial muscle cells [29] and osteoblasts [30], but to be antimitogenic in other cell types [31, 32]. Our findings that apoptosis is increased in early pregnancy and decreased near term in smokers raise the possibility that nicotine may have different effects depending on the extent of cytodifferentiation, and that the process may also be influenced by the presence of hypoxia. Although our data are supportive of a role for apoptosis in the changes observed in the placenta during the first trimester, others have not observed changes in apoptosis when exposing villi of nonsmokers to nicotine [9] and have suggested that decreased mitotic potential of cytotrophoblasts may instead be involved. Although the latter is possible, findings from these researchers do not preclude a possible involvement of apoptosis because nicotine is only one constituent of cigarette smoke.
The mechanism by which placental apoptosis becomes dysregulated during maternal smoking has not been investigated. The binding of FasL to Fas leads to receptor trimerization, activation of caspases [33], and ultimately, apoptosis. The Fas/FasL system has been shown to be of importance in immune response to viruses and tumor cells as well as in the regulation of lymphocyte homeostasis [13, 33–35]. However, its role and regulation in the placenta remain controversial and studies to date have focused primarily on its significance in the maintenance of immune privilege by the fetoplacental unit [22]. Previous studies from our laboratory have shown that although Fas expression remained relatively constant throughout placental development, a significant decrease in Xiap expression was noted at term, precisely when increased apoptosis was observed [16]. It appears therefore that under normal physiologic conditions, trophoblast apoptosis is regulated by a balance between proapoptotic and antiapoptotic proteins, and that Xiap plays a crucial role in protecting the cell against Fas-mediated apoptosis. Our current studies support this contention, because a significantly higher Xiap content was observed in smokers when compared with nonsmokers near term, at a time when decreased trophoblast apoptosis and increased proliferative activity were observed. This was in contrast with nonsmokers in whom a striking decrease in Xiap was seen, suggesting that smoking was associated with a failure of the placenta to decrease Xiap expression throughout development. In addition, because Xiap inhibits the activation of procaspase-3, a cell death protease downstream on the TNFRI and Fas pathways, which is processed during activation, it is not surprising to see an increase in procaspase-3 in smokers in the third trimester. Although our results suggest that Xiap plays a regulatory role in trophoblast apoptosis induced by tobacco smoke, they do not exclude the possibility that other cell survival factors, such as Bcl-2 and EGF receptor, may also be involved. In fact, in other tissues, including lung as well as head and neck cancers, associations between tobacco smoking and Bcl-2 have been demonstrated [36–39]. Furthermore, it has recently been demonstrated that the expression of a novel gene, T-18, correlated strongly with placental apoptosis, suggesting it may also play an important role in this process [40].
In conclusion, our data suggest that under certain pathological conditions such as fetal exposure to maternal smoking, differentiated trophoblasts have the ability to maintain expression of Xiap to protect themselves from toxic insults. The differences in the expression of proapoptotic and antiapoptotic proteins and the degree of apoptosis observed in placentae of smokers between first and third trimesters also support the hypothesis that maternal smoking has different effects at different stages of development. Our results suggest that one of the underlying molecular mechanisms involved is the regulation of trophoblast apoptosis through modulation of placental Xiap expression. We suggest that trophoblast apoptosis represents an important cellular response to the toxic effects of maternal smoking on placental growth and development.
ACKNOWLEDGMENTS
We thank Dr. R. Feigel as well as the nursing and house staff of the Ottawa Hospital for their assistance in collecting placental samples. We also thank Dr. Eric LaCasse (ApoptoGen Inc., Ottawa) for providing the Xiap antibody used in our studies.
FOOTNOTES
First decision: 27 February 2001.
1 This work was supported by a grant from the Physician Services Inc. (PSI). 
2 Correspondence: Andrée Gruslin, Division Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Room 8420, The Ottawa Hospital-General Campus, 501 Smyth Road, Ottawa, ON, Canada K1H 8L6. FAX: 613 737 8470; agruslin{at}ottawahospital.on.ca
Accepted: May 21, 2001.
Received: February 2, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  Y. Wang, P. U. Rippstein, and B. K. Tsang Role and Gonadotrophic Regulation of X-Linked Inhibitor of Apoptosis Protein Expression During Rat Ovarian Follicular Development In Vitro Biol Reprod, February 1, 2003; 68(2): 610 - 619. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
